Skip to content

Commit

Permalink
Fixed Titulky provider and made some code maintenance
Browse files Browse the repository at this point in the history
  • Loading branch information
@sambartik
sambartik committed Apr 26, 2022
1 parent a4d9eb6 commit e6551dc
Showing 1 changed file with 49 additions and 18 deletions.
67 changes: 49 additions & 18 deletions libs/subliminal_patch/providers/titulky.py
View file
Expand Up @@ -35,8 +35,6 @@

# Check if any element from source array is contained partially or exactly in any element from target array
# Returns on the first match


def _contains_element(_from=None, _in=None, exactly=False):
source_array = _from
target_array = _in
Expand Down Expand Up @@ -119,18 +117,17 @@ def get_fps(self):
def get_matches(self, video):
matches = set()
_type = 'movie' if isinstance(video, Movie) else 'episode'

# Subtitle's names (could be series/episode/movie name) present in the subtitle details page
# Consists of the main name and alternative names, stripped of the S00E00 substring
sub_names = self._remove_season_episode_string(self.names)

if _type == 'episode':
# EPISODE

# match imdb_id of a series
if video.series_imdb_id and video.series_imdb_id == self.imdb_id:
# NOTE: Is matches.add('series_imdb_id') doing anything?
# For now, let's match with the 'series' to not reject
# subs with no name but a correct imdb id.
matches.add('series')
matches.add('series_imdb_id')

# match season/episode
if self.season and self.season == video.season:
Expand All @@ -142,7 +139,7 @@ def get_matches(self, video):
if len(sub_names) > 0:
series_names = [video.series] + video.alternative_series
logger.debug(
f"Titulky.com: Finding exact match between subtitle names {sub_names} and series names {series_names}"
f"Titulky.com: Finding exact match between subtitle's names {sub_names} and series names {series_names}"
)
if _contains_element(_from=series_names,
_in=sub_names,
Expand All @@ -152,7 +149,7 @@ def get_matches(self, video):
# match episode title
episode_titles = [video.title]
logger.debug(
f"Titulky.com: Finding exact match between subtitle names {sub_names} and episode titles {episode_titles}"
f"Titulky.com: Finding exact match between subtitle's names {sub_names} and episode titles {episode_titles}"
)
if _contains_element(_from=episode_titles,
_in=sub_names,
Expand All @@ -169,7 +166,7 @@ def get_matches(self, video):
# match movie title
video_titles = [video.title] + video.alternative_titles
logger.debug(
f"Titulky.com: Finding exact match between subtitle names {sub_names} and video titles {video_titles}"
f"Titulky.com: Finding exact match between subtitle's names {sub_names} and video titles {video_titles}"
)
if _contains_element(_from=video_titles,
_in=sub_names,
Expand Down Expand Up @@ -414,8 +411,7 @@ def outer_func(*args, **kwargs):

return outer_func

# TODO: Parse name and alternative names of a series / movie
# Parse details of an individual subtitle: imdb_id, release, language, uploader, fps and year
# Parse details of an individual subtitle: imdb_id, series/movie names, release, language, uploader, fps and year
@capable_of_multithreading
def parse_details(self, partial_info, ref_url=None):
html_src = self.fetch_page(partial_info['details_link'], ref=ref_url)
Expand All @@ -439,6 +435,23 @@ def parse_details(self, partial_info, ref_url=None):
if not imdb_id:
logger.debug("Titulky.com: No IMDB ID supplied on details page.")

# SERIES/MOVIE NAMES
names = []
try:
main_name = details_container.find('h1', id='titulky').contents[0].strip()
alt_name = details_container.find('h2').contents[1].strip()
if main_name:
names.append(main_name)
else:
logger.debug("Titulky.com: Could not find main series/movie name on details page.")
if alt_name:
names.append(alt_name)
except IndexError:
raise ParseResponseError("Index out of range! This should not ever happen, but it just did. Oops.")

if len(names) == 0:
logger.debug("Titulky.com: No names found on details page.")

# RELEASE
release = None
release_tag = details_container.find('div', class_='releas')
Expand Down Expand Up @@ -529,6 +542,7 @@ def parse_details(self, partial_info, ref_url=None):

info = {
'releases': [release],
'names': names,
'language': language,
'uploader': uploader,
'fps': fps,
Expand Down Expand Up @@ -696,6 +710,7 @@ def execute_foreach(self, array, func, args=[], kwargs={}):

# Special search only for episodes. Complements the query method of searching.
def browse_episodes(self,
language,
imdb_id=None,
season=None,
episode=None):
Expand Down Expand Up @@ -747,26 +762,41 @@ def browse_episodes(self,
logger.debug("Titulky.com: No previous episode number!")
raise ProviderError("Previous episode number missing, can't parse.")

# If this row contains the first subtitles to an episode number,
# add an empty array into the episodes dict at its place.
if not last_ep_num in episodes_dict:
episodes_dict[last_ep_num] = []

details_link = f"{self.server_url}{details_anchor.get('href')[1:]}"
id_match = re.findall(r'id=(\d+)', details_link)
sub_id = id_match[0] if len(id_match) > 0 else None
download_link = f"{self.download_url}{sub_id}"
# Approved subtitles have a pbl1 class for their row, others have a pbl0 class
approved = True if 'pbl1' in row.get('class') else False

# Parse language to filter out subtitles that are not in the desired language
sub_language = None
czech_flag = row.select('img[src*=\'flag-CZ\']')
slovak_flag = row.select('img[src*=\'flag-SK\']')

if czech_flag and not slovak_flag:
sub_language = Language('ces')
elif slovak_flag and not czech_flag:
sub_language = Language('slk')
else:
logger.debug("Titulky.com: Unknown language while parsing subtitles!")

# If the language is not the desired one, skip this row
if sub_language and sub_language != language:
continue

result = {
'names': [],
'id': sub_id,
'approved': approved,
'details_link': details_link,
'download_link': download_link
}

# If this row contains the first subtitles to an episode number,
# add an empty array into the episodes dict at its place.
if not last_ep_num in episodes_dict:
episodes_dict[last_ep_num] = []

episodes_dict[last_ep_num].append(result)

# Rows parsed into episodes_dict, now lets read what we got.
Expand Down Expand Up @@ -932,7 +962,8 @@ def list_subtitles(self, video, languages):
# (0)
if video.series_imdb_id:
logger.info("Titulky.com: Finding subtitles by browsing TV Series page (0)")
partial_subs = self.browse_episodes(imdb_id=video.series_imdb_id,
partial_subs = self.browse_episodes(language,
imdb_id=video.series_imdb_id,
season=video.season,
episode=video.episode)
if (len(partial_subs) > 0):
Expand Down

0 comments on commit e6551dc

Please sign in to comment.